Compare AI models

Specifications
Parameters
1.8B/3.25B16B
Context window
32k8k
Benchmarks
MATH
22.8%45.3%
MMLU
55.8%70%
Overview
CompanyGoogleMoonshot AI
Release dateDec 6 2023Feb 22 2025
AccessClosedOpen Weight
Model detailsView modelView model

Frequently asked questions

Moonlight leads Gemini 1.0 Nano on 2 of the 2 benchmarks they both report (MATH, MMLU). Gemini 1.0 Nano shipped 444 days before Moonlight, so benchmark comparisons should account for the intervening progress.

Gemini 1.0 Nano has 1.8B/3.25B parameters, while Moonlight has 16B. Context windows are 32k (Gemini 1.0 Nano) vs 8k (Moonlight). Gemini 1.0 Nano is closed, while Moonlight is open weight.

On MATH, Moonlight leads at 45.3% vs Gemini 1.0 Nano at 22.8%. On MMLU, Moonlight leads at 70% vs Gemini 1.0 Nano at 55.8%.